commander CLAUDE.md

Project instructions for CC Commander, a guided project-management and workflow layer for Claude Code. They explain its session defaults, command structure, specialist agents, and domain-specific routing.

In plain words
What is it for?
Use them when working on CC Commander commands, skills, specialist agents, setup, deployment, rollback, updates, or its workflow and orchestration behavior.
Why use it?
They give the agent a consistent way to understand and organize work in a large Claude Code plugin with many commands and specialist roles.

Instructions file

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/kevinzai/commander/claude-md
Clone the repo
git clone --depth 1 https://github.com/KevinZai/commander
Per session 8,129 This file is loaded in full into every session.
When invoked 8,129 The same file — it is already loaded in full.
Security scan D 3 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.08129 $0.08129
Opus 5 $0.04065 $0.04065
Sonnet 5 $0.01626 $0.01626
Haiku 4.5 $0.00813 $0.00813

Measured 3d ago against content hash e930787c85cd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade D, and why

commander CLAUDE.md scanned grade D with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

├── install-remote.sh # One-line remote installer (curl | bash)

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

- **Effort:** Anthropic's Opus 5 guidance: start at `high` (the default), drop to `low`/`medium` liberally wherever evals hold quality, step up to `xhigh` only for demanding coding/agentic work, `max` only when justified

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

├── install-remote.sh # One-line remote installer (curl | bash)
CLAUDE.md · 292 lines

How it starts

The opening of the file, as written. The whole thing — 292 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — CC Commander

What This Is

CC Commander — by Kevin Zicherman. Guided AI PM to Master Claude Code Instantly. Core free forever — affiliate links + consulting + optional Pro community + GitHub Sponsors. Built on Anthropic's Claude Agent SDK sub-agent architecture — CCC is a PM/orchestration UX layer over that primitive. The Desktop plugin is the primary product — primary surface: Claude Code Desktop (aka Cowork Desktop). 83 plugin skills (13 /ccc-* specialist workflows + 11 ccc-* domain routers + 7 channel/CI/ECC/setup skills + diagnostic/meta + vendor-sourced + lifecycle + /ccc-deploy + /ccc-rollback + /ccc-onboard + /ccc-fleet-viz + /ccc-tuneup + /ccc-update + /ccc-console; the Cockpit's Browse tab separately displays a 14-domain taxonomy for filtering — a display concept, not a routers count), 22 specialist sub-agent personas (architect · reviewer · builder · security-auditor · debugger · designer · qa-engineer · devops-engineer · data-analyst · content-strategist · product-manager · performance-engineer · researcher · technical-writer · fleet-worker · typescript-reviewer · python-reviewer · go-reviewer · rust-reviewer · java-reviewer · kotlin-reviewer · csharp-reviewer), 23 lifecycle hooks × 44 handlers (SessionStart [orchestrator], UserPromptSubmit, PreToolUse, PostToolUse, Stop, Notification, PreCompact, PostCompact, SubagentStop, SubagentStart, PermissionRequest, SessionEnd, Elicitation, ElicitationResult, StopFailure, PostToolUseFailure, PostToolBatch, TaskCreated, TaskCompleted, ConfigChange, UserPromptExpansion, InstructionsLoaded, Setup), 2 credential-free bundled MCP servers (context7 + sequential-thinking) + 16 opt-in via /ccc-connect. Also a comprehensive Claude Code configuration toolkit + interactive CLI project manager: 467 skills, 11 CCC domains, 83+ commands, 24 hooks, 37 prompt templates, 10 themes, 19 vendor packages, 3 starter templates, real-time agent dashboard, OpenClaw native integration, agent-friendly CLI API, tabbed tmux split mode, status updates, continuous improvement pipeline, modular installer. Built by scanning 200+ articles from the Claude Code community and distilling into one install.

Read the full file on GitHub · 292 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 292 lines · 8,129 tokens per session scan D e930787c85cd

Subscribe to this mod's changes

commander CLAUDE.md is an instructions file published in the GitHub repository KevinZai/commander (6 stars, last pushed 3d ago), licensed MIT. It adds 8,129 tokens to every session, about $0.0406 per session on Opus 5. A static security scan graded it D with 3 findings (downloads and executes remote code, reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.